We All Agree on Vocal Anatomy, Right?

This is a long read, but some ideas are worth more than a soundbite.

Anatomy tells us what’s there. Physiology explains how it works. Function asks what the whole thing is actually accomplishing.

So vocal anatomy ought to be the easy part.

After all, we’re dealing with physical structures. Things you can dissect, photograph, scan, stain, measure and put under a microscope. Surely there are facts, and singing teachers should simply learn them.

Except it isn’t quite that simple.

Let’s take the vocal folds. They’re fairly important to singing, so you’d think we’d be on solid ground here.

How many layers does a vocal fold have?

Two.

Three.

Five.

Six.

Possibly seven.

All of those answers can be defended.

The two-layer vocal fold

One of the most influential ways of understanding the vocal folds comes from Minoru Hirano’s body-cover model.

At its simplest, the vocal fold can be thought of as having two mechanically different parts: a relatively “stiff” body and a more flexible cover.

Very useful.

If I’m teaching someone about vocal fold vibration, I may not need them thinking about collagen fibres, elastin distribution and epithelial cells. I want them to understand that different parts of the fold can behave differently during vibration. Two layers does the job.

Except, of course, they’re not literally two layers. It’s a model.

Great, that’s straightforward. Except it isn’t.

Even scientific papers describing what is usually called Hirano’s body-cover model don’t necessarily draw the boundary in exactly the same place.

One published review describes the body as the muscle plus the deep lamina propria, with the intermediate and superficial lamina propria and epithelium forming the cover.

Another contemporary scientific paper describes the two-layer version with the thyroarytenoid muscle plus both the intermediate and deep lamina propria as the body, leaving the superficial lamina propria and epithelium as the cover.

Both cite Hirano.

So even saying “according to Hirano…” may not get us quite as far as we’d like.

It doesn’t mean somebody has got the vocal fold wrong. It means a model has been interpreted, adapted and used for different purposes.

So let’s make it three

We can instead describe:

Cover
Vocal ligament
Body

The cover includes the epithelium and superficial lamina propria. The intermediate and deep lamina propria together form the vocal ligament. The muscular part forms the body.

Hirano himself also described a three-part mechanical division of cover, transition and body: epithelium plus superficial lamina propria; intermediate plus deep lamina propria; and vocalis muscle.

That’s often more useful if we’re discussing how vocal fold stiffness varies from the surface towards the muscle.

So now we have three.

Except we don’t really have three either.

Five is the proper answer. Isn’t it?

Look through the histology literature and you’ll commonly find the adult human vocal fold described as having five layers:

1. Epithelium
2. Superficial lamina propria
3. Intermediate lamina propria
4. Deep lamina propria
5. Thyroarytenoid/vocalis muscle

There we are.

Five.

We can finally write the exam.

Except between the epithelium and superficial lamina propria there’s another rather important structure called the basement membrane zone.

Do we count that?

Some descriptions explicitly include it.

Now we have six.

Unfortunately the basement membrane zone isn’t just a featureless sheet either. At finer resolution it can itself be divided into structures including the lamina lucida and lamina densa, with anchoring structures connecting the epithelium to the tissue beneath.

So if we’re going to be properly anatomically accurate, perhaps we’re heading towards seven.

And while we’re doing that, the epithelium itself isn’t a single row of cells. Vocal fold epithelium is stratified squamous epithelium, several cells thick.

At this point the perfectly reasonable question “How many layers are there in a vocal fold?” has become surprisingly difficult to answer.

The problem isn’t that anatomy is wrong.

The problem is the question.

A model isn’t the same thing as a structure

This distinction gets lost surprisingly easily in voice teaching.

The two-layer body-cover model and the five-layer histological description aren’t competing accounts where one person has got their anatomy wrong. They’re looking at the same tissue at different levels of organisation, for different reasons.

One describes mechanical behaviour.

One creates functional groupings.

One describes histology.

Zoom in further and you’ll get cells, collagen, elastin, hyaluronic acid, extracellular matrix, proteins and molecular structures.

Zoom out and you’ve got a vocal fold.

None of these descriptions has to cancel out the others. They’re maps at different scales.

The London Underground map isn’t geographically accurate. It’s an extremely useful map because it leaves out information we don’t need. Nobody therefore concludes that Harry Beck didn’t understand London.

Voice anatomy teaching works in much the same way.

But voice pedagogy has another problem

We don’t just borrow from anatomy. Voice pedagogy is an extraordinary magpie discipline.

We pick things up from anatomy, histology, physiology, acoustics, biomechanics, neuroscience, psychology, motor learning, speech science, medicine, linguistics, education and probably half a dozen other fields I’ve forgotten.

And then we add artistry.

That’s one of the things I love about teaching voice. It’s also one of the things that makes it dangerous to talk about “the science” as though it were a single coherent body of instructions for singing teachers. Every one of those disciplines asks different questions.

A histologist may be interested in what tissues are present.

A biomechanist may want to know how those tissues deform under load.

An acoustician may be measuring the resulting signal.

A neurologist may be investigating control.

A speech scientist may be studying ordinary spoken phonation.

A clinician may be interested in pathology.

A motor-learning researcher may be studying how people acquire a task.

And a singer may simply be trying to get through the second act of Wicked eight times a week.

Those aren’t the same question.

Yet something quite subtle can happen when information travels from one field into voice pedagogy.

The context falls off.

The disappearing context problem

A paper studies a particular group of people, performing a particular task, under particular experimental conditions. A result is found. The researchers discuss what they think that result might mean, usually with a section devoted to limitations and questions that remain unanswered.

Then somebody reads the abstract.

Or hears the researcher speak at a conference.

Or sees one slide.

Or reads somebody else’s description of the paper.

And before long:

“Researchers found…”

becomes

“Science has proved…”

and eventually

“Therefore singers should…”

Those are three very different statements. So we start to ask…

• What population was studied?
• Speaking or singing?
• Trained singers or untrained participants?
• Classical singers, musical theatre performers, CCM singers?
• How many people?
• What task?
• At what pitches and intensities?
• What was actually measured?
• Was the study describing an association, proposing a mechanism or demonstrating causation?
• Was it an experiment, a modelling study, a clinical observation, an expert hypothesis or a theoretical paper?

And, crucially, did the researchers actually make the pedagogical claim that is now being attributed to them? Often they didn’t.

“Voice scientist X says…”

This is another one I see surprisingly often. A voice scientist’s name gets attached to an idea as a kind of quality stamp.

“Professor X says this happens.”

“Dr Y proved that.”

“This technique is backed by the research of…”

Sometimes I know the research being referred to. And occasionally I know the researcher. And I find myself thinking:

That’s not quite what they said.

They may have measured something under particular experimental conditions. They may have proposed one possible explanation for what they observed. They may even have explicitly warned against making the very leap that is now being made in their name.

But once a sentence has been detached from its context, it becomes wonderfully portable.

It can turn into a diagram.

Then a PowerPoint slide.

Then a teacher-training handout.

Then an Instagram post.

Then, eventually, a Fact About The Voice.

At which point the original scientist may barely recognise it.

This isn’t necessarily dishonesty. Most of the time I suspect it’s much more ordinary than that.

We like certainty.

Teachers need explanations they can use. Students like answers. Social media rewards statements rather more enthusiastically than qualifications.

“This may be one useful way of interpreting these findings, although there are several limitations” isn’t quite as catchy as:

“SCIENCE SAYS YOU’RE BREATHING WRONG.”

But intellectually, the first sentence may be much closer to the truth.

And then there’s artistry

Singing adds another complication.

Some of the most important knowledge in our profession doesn’t originate in a laboratory at all. It comes from skilled observation, experience, experimentation, tradition, imagination, sensation, communication and artistry.

That doesn’t automatically make it unscientific nonsense.

And adding an anatomical explanation doesn’t automatically make it more true.

A singing teacher can discover an extraordinarily effective way of helping someone produce a sound without knowing the precise physiological mechanism responsible.

Conversely, a teacher can give a beautifully detailed anatomical explanation which is completely useless to the person trying to sing the phrase.

Those are different kinds of expertise.

I’m interested in how they inform each other. I’m much less interested in using one to outdo another in status.

Physiology gets even messier

Anatomy at least has the advantage that we’re looking at things that physically exist.

Physiology requires us to explain what those structures are doing.

And function asks why.

Now we are interpreting movement, forces, pressures, neural activation, tissue deformation and acoustic consequences, often involving structures we cannot directly observe while somebody is singing.

Which is how a useful explanation can gradually turn into a “fact”.

“The cricothyroid stretches the vocal folds.”

Useful shorthand.

“The thyroarytenoid shortens the vocal folds.”

Also useful.

But neither tells you everything either muscle is doing, how activation changes vocal fold geometry, how the two muscles interact, or what happens when several intrinsic and extrinsic muscles are active simultaneously.

The sentence isn’t necessarily wrong.

It’s incomplete.

And incomplete information is unavoidable in teaching.

This is why language matters

I don’t want singing teachers frightened of talking about anatomy in case they get a microscopic detail wrong. I want them to get better at signalling what kind of statement they’re making.

There’s an enormous difference between:

“The vocal folds have two layers.”

and

“For understanding vibration, it can be useful to model the vocal folds as a body and a cover.”

Or between:

“The vocalis does X.”

and

“One way of describing the medial portion of the thyroarytenoid is as the vocalis, and activation in this region contributes to…”

Or between:

“This is what happens.”

and

“This model helps us explain what we’re seeing/feeling.”

Between:

“Research proves…”

and

“In this particular study, researchers found…”

Those tiny changes in language make teaching more accurate without making it incomprehensible.

They also leave room for the science to change.

Which it does.

The qualification problem

And this is where I struggle with the idea that we can solve the problem by creating an anatomy qualification that proves singing teachers know the facts.

• Which facts?
• At what level?
• According to which textbook?
• Which model?
• How recent does the research have to be?
• Would “five vocal fold layers” get full marks?
• Would “body and cover” be marked wrong?
• Which version of body and cover?
• Would someone mentioning the basement membrane get extra credit?
• What happens when two perfectly respectable scientific papers use slightly different classifications?

And perhaps we need another set of questions.

• Can the teacher distinguish a histological finding from a biomechanical model?
• Can they tell the difference between research on speech and evidence about singing?
• Can they read past the abstract?
• Can they separate what researchers actually found from what the researchers think it might mean?
• Can they notice when somebody else has made a much bigger claim than the original study did?
• Can they say “we don’t know”?

A multiple-choice anatomy test may tell me whether somebody has memorised the terminology chosen by whoever wrote the test.

I’m not convinced it tells me whether they’re a better singing teacher.

I’d be much more interested in whether they understand the limitations of the information they’re using.

Can they distinguish anatomical observation from a functional model?

Can they explain something simply without turning the simplification into immutable fact?

Can they notice when two apparently contradictory descriptions are actually looking at the same system from different perspectives?

Can they change their mind when better evidence appears?

Can they hold scientific knowledge, practical experience and artistic knowledge in the same room without pretending they’re interchangeable?

And perhaps most importantly, do they know when their anatomical explanation is actually useful to the singer standing in front of them?

Know your anatomy. And know what kind of knowledge it is

None of this is an argument for ignorance. It’s the opposite.

The more anatomy and physiology I’ve learned over the years, the less interested I’ve become in statements beginning “We now know that…”

We usually know something much more interesting.

We know that under particular conditions, using a particular method, researchers observed something that supports a particular interpretation.

That’s not weakness. That’s science.

So yes, I want singing teachers to understand anatomy and physiology.

I want them curious about acoustics, biomechanics, motor learning and all the other subjects that feed our peculiar profession.

I want them reading research, questioning diagrams, comparing sources, checking quotations and updating what they teach.

I want them to be suspicious when a complicated idea becomes unexpectedly certain after passing through several people.

And I want them to value the knowledge that comes from years spent actually listening to and working with singers without feeling obliged to dress every useful artistic insight in a lab coat.

Perhaps scientific literacy for a singing teacher isn’t being able to give the approved answer to:

“How many layers does a vocal fold have?”

Perhaps it’s knowing enough to reply:

“How detailed would you like me to be?”

 

 

If you enjoyed this blog, we show you how to sort complex issues and ask the right questions in our new Teacher-Mentor Pathway – check it out.